List your company

Sycophancy detection on LURE Truthfulness

10Sycophancy RateGemini 3 Flash
8.0821.043446.96Apr 8, 2026

Evaluation results

MethodLinks
Gemini 3 FlashFamily=Gemini, Organization=Google2026.041058
Gemini 3 ProFamily=Gemini, Organization=Google2026.041142
GPT-5.2Family=GPT, Organization=OpenAI2026.041835
GPT-4oFamily=GPT, Organization=OpenAI2026.042212
GPT-4.1Family=GPT, Organization=OpenAI2026.042622
DeepSeek V3Organization=DeepSeek2026.043611
Sonnet 4.5Family=Claude, Organization=Anthropic2026.044436
Opus 4.6Family=Claude, Organization=Anthropic2026.044644
Llama 4 Maverick†Family=Llama, Organization=Meta, Outlier=Discussed in text2026.04553
Llama 4 ScoutFamily=Llama, Organization=Meta2026.045842